Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Kagus]

Wish I could get that in Scrabble...


In RPGs today, there are a number of ways of getting characters to progress.  The most common variant is standard EXP and levels, you do (kill) things, get points, and use those points to earn levels.  Another method is purely equipment-based, in that your character has no or very little intrinsic ability, but the things they can collect and wear on their heads makes them stronger and capable of getting even prettier hats in different colors.

One of my favorites, however, is the fluid skill training system.  Like our very own Dwarf Fortress, you have to do things to get better at them.  Dungeon Siege, Illarion, heck even Runescape use this method.  Every time you use a skill, you get better at using it.  Makes sense.


However, after seeing some truly dreadful implementations of this particular system, and some of the problems that come from them, I've come to wonder if maybe there's a better way of doing things.

In standard skill training systems, whenever you correctly perform an action, the associated skill either has a chance to increase, or does increase by an incremental value.  One of the peculiar things about this is that the completely uninitiated will have a terrible time learning due to their inability to successfully do that thing even once.  The more experienced, the more you learn, which seemed a bit odd to me.

What if, instead, skill gains came from failing at an action?  (This is where Nietszche comes in to make things sound fancy, "That which does not kill me, only makes me stronger")

Humans in reality often learn from mistakes...  Or, rather, we learn a fair sight more from our mistakes than from our successes, generally speaking.  If we were to attempt to implement this in a game, we'd potentially see a system where someone with no knowledge of a skill (say, swordfighting) would have no real concept of how things should be done, and would thus stand to learn the most.  As they take a few practice fights and earn a few scars and bruises, they begin to develop an understanding for how the opponent can and will move to attack them (thus opening the way to learn how to defend against them, through trial and error), and how to wield the weapon in order to better put the opponent offguard, to wound, and eventually to kill.  They would, through their many failings, quickly learn the basics of swordplay and develop a rudimentary grasp of the concept in very short time, while refining the skill would take more practice, and more enlightenment as to what can be refined.

Let's take our swordsman now and fast forward a bit...  He's been hunting goblins for years and years, and knows everything about their combat style.  He can perfectly dodge, parry and avoid their attacks, and knows exactly how and where to strike to gain the most efficacy from his movements.  He is, at this point, perfect.  And stands to learn nothing more.  The pattern merely cements itself through practice, he will continue to be perfect at slaying goblins.  He can't get any "more perfect", because he's not learning anything from this, he's just doing what he knows best.

Now, pit him against a seasoned knight, who has spent his life adventuring in far-off lands and crossing blades with far stranger and dangerous foes.  Suddenly, our goblinslayer is pitted against a greater swordsman with new and entirely different tactics, and our swordsman finds his knowledge to be insufficient.  But, again, learning from his failures to divert the knight's blade or penetrate his defenses, he soon grows to have a greater understanding of swordplay and matches the knight's skill.


This, I feel, makes more sense.  It makes new skills easy to learn but difficult/dangerous to perfect, and performing repetitive, menial and "safe" actions will not serve to make you any better, because you're not challenging yourself.  This forces players to seek out greater risks and challenges, or even impose restrictions on themselves ("I could take you all on with one hand tied behind my back!" ... "I am not left-handed!") in order to attain a higher degree of skill.

This also makes grinding, in some cases, a bit less spammy.  I can recall all too clearly all the molotovs I had in HellMOO, all the wooden plates and iron daggers that were constructed and sold en masse in Illarion...  It just serves to produce a vast amount of junk that nobody knows what to do with afterwards.  Rather, if you learn by not getting it right, you don't get a vast amount of finished products lying around.  In contrast, it also means producing for the sake of the product has to be done exactly for that reason, since you probably aren't getting any skill boosts out of it.  And, as a consolation, every failure to make something helps you get better at making it next time.



Now, this idea obviously has its own flaws...  One could simply grind away at the most complex, difficult task they could find and skill up that way.  But it seems simple enough to make such actions fairly dangerous, or at least rather expensive.  So, again, not impossible...  Which it shouldn't be.  Just discouraged.


Anyone care to voice their thoughts on the matter?  Are there actually any titles out there that do something similar to this?  Would you like to see this system in a game, and how do you think it could be improved or altered?

[Blaze]

In 3079, 3 skills (our of 4) were only trained upon failing. This was abusable by using items that decreased your skill, allowing you to easily gain lots of skill points.

I'd say one of the better implementations of the skill comes from the Mount&Blade Warband Mod: Rigale. Since skills determine the skill roll, all the skill gains are roll-based. If you easily succeed, you gain no skill xp; but if you just barely succeeded, you'd gain more XP than when facing an easier skill roll. On the other end of the spectrum; if you failed, and had enough INT to understand why you failed, you'd gain some XP, if you somehow failed epically, you'd gain no XP.

[Ozyton]

Kenshi is supposed to make it so failure gives you more experience than winning, to sort of make it so you don't just savescum all the time. Failure of course has its own problems...

Don't quite know if it's implemented yet, but it's planned.

EDIT: This is why most enemies in games give a very small amount of XP if they're low level but higher XP if they're higher level. A low level guy killing low level creatures are going to get decent XP for their level since they're fighting a fair fight, but if they manage to take out a significantly harder opponent they will get a ton of XP since they would have 'learned more' from that particular fight. Some games make it so that the amount of XP you get depends on the level difference, so a level 50 guy might get 10 xp from a level 20 enemy, but a level 10 guy would get a lot more than 10 xp for taking out the same enemy. It balances the game a little. It's not the perfect system, but it's a system that has worked for a while.

[frostshotgg]

The problem with this set-up, and the normal practice-makes-perfect system, is that it doesn't alleviate grinding. If anything, this system makes the problem worse, because every time you're successful, you have to make more tries to fail.

[Microcline]

As much as Bethesda and other companies tout "you are what you play" skill systems, I've never really found an implementation that works well.  It always comes back to the fact that tends to reward repetitive meta-gaming and skill spamming over normal play.  I don't know where the hostility to traditional experience + point buy systems comes from, as long as experience is awarded for actions such as exploring new locations or completing quests like in New Vegas instead of repetitive murder (apart from the fact that "level-less", "class-less", and "innovative skill system" have become toxic marketing blurbs).

There's also the fact that such a system would barefacedly reward failure and punish success, which is something to avoid at all costs.

[PrimusRibbus]

For your average RPG, I've come back around to XP and levels. I really like "skill by doing" systems on paper (and it works great in games like DF), but for many games I've played it turns into pure grind in practice.

Non-combat skills, in particular, tend to become pointless grindfests when exposed to the "skill by doing" system.

[Trollheiming]

Maybe a little off-topic, but I never got the quote from Nietzsche. If someone cuts off my arms and legs, rips out my tongue, and pokes out my eyes, I really don't feel as though I've been strengthened. Even mentally, I'd probably just go insane.

[BuriBuriZaemon]

I have been reading about the application of "learning from failures" in business perspective. In essence, while failures may lead to the acquisition of new skills or significant improvement of current skills, they carry a cost (e.g. brand damage or monetary losses).

In game designs, I imagine this is easily applicable in crafting-based skills or any skills that involve the usage of tools. Failures in using these tools or crafting materials lead to them being damaged/lost (as a cost to the player) but at the same time, gives greater increment in the player's skill progression than successful outcomes.

[Ozyton]

Maybe a little off-topic, but I never got the quote from Nietzsche. If someone cuts off my arms and legs, rips out my tongue, and pokes out my eyes, I really don't feel as though I've been strengthened. Even mentally, I'd probably just go insane.

I've heard that if you break a bone it heals and becomes stronger than before. I think tribes in Africa or something actually break their people's bones and stuff for that reason.

That's a little besides the point though XD

[MorleyDev]

Maybe a little off-topic, but I never got the quote from Nietzsche.

He was specifically talking about wise and successful people when he said this, not making a general statement. A person who succeeds is one who can turn their own disadvantages into advantages: So long as they live through it, they have the wisdom to use it to further themselves. It's from Ecce Homo: http://en.wikipedia.org/wiki/Ecce_Homo_(book)

It's the last book he wrote before going completely insane (The Will To Power being put together by his sister from his notes, and suspected to have been highly edited by her to conform to her Nazi beliefs) and an attempt at a biography.

[frostshotgg]

I've heard that if you break a bone it heals and becomes stronger than before

That's just an old wives' tale. Any doctor worth his salt would laugh at you for saying that.

[Blaze]

In the end I just prefer the Point-Buy system. Sure, killing wolves to gain sewing skill doesn't make sense or get rid of grinding; but if you prefer action A over action B what would you be doing? This way I get to be EXACTLY what I choose to be, without doing the stuff I don't find fun. Of course, balancing said actions to give appropriate amounts of points tends to be terribly difficult as likes/dislikes differs between players.

This is especially important in games with moronic features like broken level scaling. I'm looking at YOU Bethesda, Skyrim fixed NOTHING.

[alexandertnt]

I think the issue with grinding is why your gaining that level in the first place, that is the rest of the game. If the game requires it to progress, then yes people are going to have to grind one way or another. If the game's main focus is "rewarding" the player, then people are going to find a way to grind one way or another.

It also depends on the player. Some people are "in it to win it" and some people just like gaining levels. These people will find a way to optimally grind on your system no matter what you implement.

(Using smithing as an example)

I much prefer the skill-by-doing system over the xp-and-level system because it just seems silly to increase your smithing level by smacking goblins over the head with a club. It really doesnt make my character seem like they got better at smithing. Ontop of that the mechanics behind it are usually so simple they dont entertain at all. Gaining a level is just increasing an integer. Meh.

The idea of learning from failures is appealing to me, as it seems more natural/plausable (avoiding the word realistic intentionally) than what games currently do. Perhaps it can be spiced up a bit more though, instead of simply printing the word FAILURE above the characters head in yellow text (This is how I picture it), you actually "succeed" each time, even smithing some super high level armor. However, at low skill you get a really shitty quality armor. Throw in a bit of randomness (you might get something okish, or even something more interesting, like a variety of effects (weighs you down more because its stiff and poorly made, makes using weapon X easier because you forgot to smith that restrictive part of the armor, but protection sucks etc)). This way you still actualy get something for your work. This could also make "failure" (crappy armor) a reasonably interesting game mechanic, instead of automatically dismissing the result as "yet another failure".

Furthermore to make it seem more dynamic, the game could not just quantify the smithing skill to an integer, but perhaps only express that you are "good" at smithing, or "terrable" at smithing. Or even not say anything and let the player work out how good they are at smithing by actually smithing and observe the result.

Perhaps this interests me because im more interested in poking around with game mechanics, experementing and prefer to have a more "fuzzy" understanding of my character's skills, rather than a direct readout. Basically IMO games need to work on making the act of getting better at a skill entertaining, to drive the player's desire to smith through being a fun and interesting game mechanic, instead of being a perpetual chase for the next level.

In the end I just prefer the Point-Buy system. Sure, killing wolves to gain sewing skill doesn't make sense or get rid of grinding; but if you prefer action A over action B what would you be doing

But if you prefer action A over action B, and dont enjoy action B, why would you want to increase the ability to do action B at all? Action B is an example of something that should be made reasonably fun to do, because if people are killing wolves to level their sewing skill, then the game probably implemented a boring and uninteresting sewing skill (using sewing as an example, replace with whatever you want). (Here im speaking from a game-design point of view, not saying your or any other player is wrong here or anything).

[Blaze]

But if you prefer action A over action B, and dont enjoy action B, why would you want to increase the ability to do action B at all? Action B is an example of something that should be made reasonably fun to do, because if people are killing wolves to level their sewing skill, then the game probably implemented a boring and uninteresting sewing skill (using sewing as an example, replace with whatever you want). (Here im speaking from a game-design point of view, not saying your or any other player is wrong here or anything).

It's a matter of player skill vs in-game skills mostly. Killing wolves give pelts, and pelts can be used for stuff to make sure we can keep going killing wolves. If we did indeed make sewing interesting (A minigame maybe) and still disliked it, we'd buy points to make it go by faster. If we enjoy killing wolves; why put points in it to make it easier, thus shortening the time we spent doing it?

Conversely, if we enjoyed sewing, we'd take the experience we gained from sewing into combat so more time could be spent sewing.

It may not make sense, but neither does going around killing wolves, knitting socks, and casting restoration spells while jumping in order to increase those necessary skills.

[PrimusRibbus]

snip

A few years back I would've agreed with you 100%. I really do believe, though, that developers have worried about the plausibility of their mechanics so much that they've sacrificed good game design. Sometimes real life logic just doesn't make for good video games.

But if you prefer action A over action B, and dont enjoy action B, why would you want to increase the ability to do action B at all? Action B is an example of something that should be made reasonably fun to do, because if people are killing wolves to level their sewing skill, then the game probably implemented a boring and uninteresting sewing skill (using sewing as an example, replace with whatever you want). (Here im speaking from a game-design point of view, not saying your or any other player is wrong here or anything).

Doing something a few times and having it matter is a rewarding experience. Having to button-mash 300 times just to be allowed to be better at it is pure grind.

A good example is Smithing in Skyrim. Gathering the rare materials to make an impressive weapon was rewarding; having to grind hundreds of junk items just to be allowed to make that weapon was not rewarding.
A bigger question is whether Smithing (action B) was ever intended to be a fun activity in and of itself. I'd say no; the fun of Smithing was that it supported combat (action A).